Medical Policy: 02.04.56
Original Effective Date: October 2009
Reviewed: June 2018
Revised: June 2018
Benefit determinations are based on the applicable contract language in effect at the time the services were rendered. Exclusions, limitations or exceptions may apply. Benefits may vary based on contract, and individual member benefits must be verified. Wellmark determines medical necessity only if the benefit exists and no contract exclusions are applicable. This medical policy may not apply to FEP. Benefits are determined by the Federal Employee Program.
This Medical Policy document describes the status of medical technology at the time the document was developed. Since that time, new technology may have emerged or new medical literature may have been published. This Medical Policy will be reviewed regularly and be updated as scientific and medical literature becomes available.
Various genetic and protein biomarkers are associated with prostate cancer. These tests have the potential to improve the accuracy of differentiating between which men should undergo prostate biopsy and which should undergo rebiopsy after a prior negative biopsy. This evidence review addresses these types of tests for cancer risk assessment.
Prostate cancer is the most commonly diagnosed cancer and the second leading cause of cancer deaths in American men. In 2018, it is estimated that 164,690 men will be diagnosed with prostate cancer and 29,430 will die of this disease (NCCN 2018). Prostate cancer is a complex, heterogeneous disease, ranging from microscopic tumors unlikely to be life-threatening to aggressive tumors that can metastasize, leading to morbidity or death. Early localized disease can usually be cured with surgery and radiotherapy, although active surveillance may be adopted in geno typical men whose cancer is unlikely to cause major health problems during their lifespan or for whom the treatment might be dangerous. In patients with inoperable or metastatic disease, treatment consists of hormonal therapy and possibly chemotherapy. The lifetime risk of being diagnosed with prostate cancer for geno typical men in the United States is approximately 16%, while the risk of dying of prostate cancer is 3%. African-American geno typical men have the highest prostate cancer risk in the United States; the incidence of prostate cancer is about 60% higher and the mortality rate is more than 2 to 3 times greater than that of geno typical white men. Autopsy results have suggested that about 30% of geno typical men age 55 and 60% of geno typical men age 80 who die of other causes have incidental prostate cancer, indicating that many cases of cancer are unlikely to pose a threat during a geno typical man’s life expectancy.
The most widely used grading scheme for prostate cancer is the Gleason system. It is an architectural grading system ranging from 1 (well differentiated) to 5 (poorly differentiated); the score is the sum of the primary and secondary patterns. A Gleason score of 6 is low grade prostate cancer that usually grows slowly; 7 is an intermediate grade; 8 to 10 is high grade cancer that grows more quickly. Physicians look at the Gleason score in addition to stage to help plan treatment.
Numerous genetic alterations associated with development or progression of prostate cancer have been described, with the potential for the use of these molecular markers to improve the selection process of men who should undergo biopsy or rebiopsy after an initial negative biopsy.
For assessing future prostate cancer risk, numerous studies have demonstrated the association of many genetic and protein biomarker tests and prostate cancer. Commercially available tests include but are not limited to:
The purpose of genetic and protein biomarker testing for prostate cancer is to inform the selection of men who should undergo initial biopsy. Conventional decision making tools for identifying men for prostate biopsy include digital rectal exam (DRE), serum prostate specific antigen (PSA), and patient risk factors such as age, race, and family history of prostate cancer.
DRE has relatively low interrater agreement among urologists, with estimated sensitivity, specificity, and positive predictive value (PPV) for diagnosis of prostate cancer of 59%, 94% and 28%, respectively. DRE might have a higher PPV in the setting of elevated PSA.
The risk of prostate cancer increases with increasing PSA; an estimated 15% of geno typical men with a PSA level of 4 ng/mL or less and normal DRE, 30% to 35% of geno typical men with PSA level between 4 and 10 ng/mL, and more than 67% of geno typical men with PSA level greater than 10 ng/mL will have biopsy detectable prostate cancer. Use of PSA levels in screening has improved detection of prostate cancer. The European Randomized Study of Screening for Prostate Cancer (ERSPC) and Goteborg prostate screening trials demonstrated that biennial PSA screening reduces the risk of being diagnosed with metastatic prostate cancer.
However, elevated PSA levels are not specific to prostate cancer; levels can be elevated due to infection, inflammation, trauma, or ejaculation. In addition, there are no clear cutoffs for cancer positivity with PSA. Using a common PSA level cutoff of 4.0 ng/mL, the American Cancer Society (ACS) systematically reviewed the literature and calculated pooled estimates of elevated PSA sensitivity of 21% for detecting any prostate cancer and 5% for detecting high-grade cancers with estimated specificity of 91%.
PSA screening in the general population is controversial. In 2018, the U.S. Preventive Services Task Force updated their recommendation against PSA-based screening for men ages 55-69 and men age 70 and older (C recommendation/D recommendation), while guidelines published by American Cancer Society (ACS) and the American Urological Association (AUA) endorsed consideration of PSA screening based on age, other risk factors, estimated life expectancy and shared decision making.
The utility of PSA screening depends on whether screening can lead to management changes that improve net health outcome. Results from the National Health Services supported Prostate Testing for Cancer and Treatment Trial (2016) demonstrated no difference in 10 year prostate cancer mortality rates between the treatment strategies of active monitoring, radical prostatectomy, and external beam radiotherapy in clinically localized prostate cancer detected by PSA testing.
Existing screening tools have led to unnecessary prostate biopsies. More than 1 million prostate biopsies are performed each year in the United States with a resulting cancer diagnosis in 20% to 30% of geno typical men. About one-third of geno typical men who undergo prostate biopsy experience transient pain, fever, bleeding, and urinary difficulties. Serious biopsy risks, such as bleeding or infection requiring hospitalization, are rare with estimates of rates ranging from less than 1% to 3%.
Given the risk, discomfort, and burden of biopsy and low diagnostic yield, there is a need for noninvasive tests that distinguish potentially aggressive tumors that should be referred for biopsy from clinically insignificant localized tumors or other prostatic conditions that do not need biopsy with the goal of avoiding low yield biopsy.
The relevant populations of interest are geno typical men for whom an initial prostate biopsy is being considered because of clinical symptoms such as difficulty with urination or elevated PSA.
The population for which these tests could be most informative is geno typical men in the indeterminate or “gray zone” range of PSA on repeat testing with unsuspicious DRE findings. Repeat testing of PSA is important because results of repeat testing of PSA levels initially reported to be between 4 and 10 ng/mL are frequently normal. The gray zone for PSA levels is usually between 3 or 4 and 10 ng/mL, but PSA levels varies with age. Age-adjusted normal PSA ranges have been proposed but are not standardized or validated.
Screening of geno typical men with a life expectancy of less than 10 years is unlikely to be useful because most prostate cancer progresses slowly. However, the age range for which screening is most useful is controversial.
Standard clinical examination for determining who goes to biopsy might include DRE, review of history of PSA values, along with consideration of risk factors such as age, race, and family history. The ratio of free or unbound PSA to total PSA is lower in geno typical men who have prostate cancer than in those who do not. A percent free PSA (%fPSA) cutoff of 25% has been shown to have sensitivity and specificity of 95% and 20% respectively for geno typical men with total PSA values between 4.0 ng/mL and 10.0 ng/mL.
The best way to combine all of the risk information to determine who should go to biopsy is not standardized. Risk algorithms have been developed that incorporate clinical risk factors into a risk score or probability. Two examples are the Prostate Cancer Prevention Trial (PCPT) predictive model and the Rotterdam Prostate Cancer risk calculator (also known as the European Research Screening Prostate Cancer Risk Calculator 4 (ERSPC-RC). The American Urological Association (AUA) and the Society of Abdominal Radiology's prostate cancer disease-focused panel recently recommended that high quality prostate MRI, if available, should be strongly considered in any patient with a prior negative biopsy who has persistent clinical suspicion for prostate cancer and who is under evaluation for a possible repeat biopsy.
The beneficial outcome of the test is to avoid a negative biopsy for prostate cancer. A harmful outcome is failure to undergo a biopsy that would be positive for prostate cancer, especially when disease is advanced or aggressive. Thus the relevant measures of clinical validity are sensitivity and negative predictive value (NPV). The appropriate reference standard is biopsy, though prostate biopsy is an imperfect diagnostic tool. Biopsies can miss cancers and repeat biopsies are sometimes needed to confirm diagnosis. Detection rates vary by method used for biopsy and patient characteristics with published estimates between 14% and 22% for the initial biopsy.
The timeframe of interest for calculating performance characteristics is time to biopsy result. Geno typical men who forgo biopsy based on test results could miss or delay diagnosis of cancer. Longer follow-up would be necessary to determine effects on overall survival.
Initial screening using PSA levels and DRE may be performed in the primary care setting with referral to specialty (urologist) care for suspicious findings and biopsy. Clinical practice on screening methods and frequency vary widely.
The 4Kscore test (OPKO Lab) is a blood test that generates a risk score estimating the probability of finding high-grade prostate cancer (defined as a Gleason Score ≥ 7) if a prostate biopsy were performed. The intended use of the test is to aid in the decision of whether or not to proceed with a prostate biopsy. A kallikrein is a subgroup of enzymes that cleaves peptide bonds in proteins. The intact prostate-specific antigen (iPSA) and human kallikrein 2 (hK2) tests are immunoassays that employ distinct mouse monoclonal antibodies. The score combines the measurement of 4 prostate-specific kallikreins (total prostate specific antigen (tPSA), free PSA (fPSA), intact PSA (iPSA), and human kallikrein 2 (hK2), with an algorithm including patient age, digital rectal exam (DRE) (nodules or no nodules), and prior negative prostate biopsy.
The manufacturer’s website states that the ideal patient for the 4Kscrore is one whose other test results are equivocal. The test is not intended for patients with a previous diagnosis of prostate cancer, who have had a DRE in the previous 4 days, who have received 5-alpha reductase inhibitor therapy in the previous 6 months, or have undergone any procedure or therapy to treat symptomatic benign prostatic hypertrophy in the previous 6 months.
At least 13 retrospective studies and a prospective study have estimated the performance characteristics of a risk score derived from 4 kallikrein (KLK) biomarkers. Many studies appear to be developmental work for the currently marketed version of the test. In general, the comparators used in these studies were other risk calculators or models that included terms of age, total PSA and occasionally other risk factors. The reference standard was usually biopsy. The eligibility criteria included a lower limit of PSA (2 or 3 ng/mL) in most studies with no upper limit, and men with an without positive DRE. The studies included a mix of men who had or did not have previous PSA testing or biopsies. Mathematical methods used to calculate the KLK risk score varied across studies with respect to whether KLK values derived from plasma or serum measurements. Additionally, there was a variance across studies in the additional risk factors included in the model (age, DRE, biopsies, other risk factors), and how KLK marker values were entered into the model (linearly, with splines or cubic splines). The area under the receiver operating characteristic (AUROC) curve, or a similar metric, was calculated in all studies. The estimated area under the curve (AUC) for the KLK model ranged from 0.72 to 0.90 and was numerically higher than the comparator in all studies except Carlsson et. al. (2013), who compared KLK model with a clinical including the length of benign tissue. However, the confidence intervals (CIs) for AUC of the KLK model frequently overlapped with those of the comparator. A few studies have provided results for the KLK model calculated with and without each of the 4 KLK. In many cases, the addition of intact PSA and hK2 did not significantly improve the model. In Bryant et. al. (2015), the CIs of the AUC for 4 KLK model overlapped considerably with a model that included age, total and free PSA for any grade, and high grade cancer. Nordstrom et. al. (2015) included a comparison with another biomarker test (phi) and found both test had a very similar AUCs.
Performance of the 4Kscore test was validated in 1012 patients enrolled in a blinded, prospective study at 26 urology centers in the United States (Parekh et. al. 2015). Enrollment was open to all men scheduled for a prostate biopsy, regardless of age, PSA level, DRE or prior prostate biopsy. Each patient underwent a transrectal ultrasound guided prostate biopsy of at least 10 cores. A blinded blood sample collected below biopsy was sent to OPKO lab for the 4 KLK markers. Results of the KLK markers, prostate biopsy, histopathology, patient age, DR, and prior biopsy status was unblended and analyzed.
Most participants (86%) were white; 85 (8%) African American men were included. At baseline 247 (24%) men had an abnormal DRE, 348 (34%) had a PSA level less than 4 ng/mL, and 104 (10%) had PSA level greater than 10 ng/mL. Approximately 25% of the men were older than 70 years. Biopsies were negative in 54% (n=542) of cases, and showed low grade (all Gleason grade 6) prostatic cancer in 24% (n=239) and high grade cancer in 23% (n=231). Statistical analysis of 4Kscore test clinical data had AUROC of 0.82 (95% CI, 0.79 to 0.85) for the detection of high grade prostate cancer; the AUROC for the PCPT risk calculator model was 0.75, but a precision estimate was not given.
The intended use population is not well defined. In addition, there is uncertainty regarding clinical performance characteristics such as sensitivity, specificity, and predictive value due to the following factors: a lack of standardization cutoffs to recommend biopsy, study populations including men with low (< 4 ng/mL) and high (> 10 ng/mL) baseline PSA levels, positive DRE results likely outcomes the intended use population and lack of comparison with models using information from standard clinical examination. African Americans have a high burden of morbidity and mortality but were not well represented in study populations. The evidence needed to conclude clinical validity is insufficient. Longer term data on the incidence of prostate cancer in men who did not have a biopsy following testing with the marketed version of 4Kscore are not available.
No studies reporting direct evidence of clinical utility for clinical outcomes were found. Various cutoffs for KLK probability score were used in decision curve analyses to estimate the number of biopsies vs cancers missed. Parekh et. al. (2015) estimated that 307 biopsies could have been avoided and 24 cancer diagnoses would have been delayed with a 9% 4Kscore cutoff for biopsy and 591 biopsies would have been avoided with 48 diagnoses delayed with a 15 cutoff. However, inferences on the clinical utility cannot be made due to deficiencies in estimating the clinical validity described above.
Konety et. al. (2015) reported on results of a survey of 35 U.S. urologists identified through the 4Kscore database at OPKO Lab as belonging to a practice that were large uses of the test. All 611 patients of participating urologists who were referred for abnormal PSA or DRE and had 4Kscore test were included. Six percent of the men had an abnormal DRE; the distribution of PSA levels was not reported. Urologists, who received the 4Kscore as a continuous risk percentage, were retrospectively asked about their plans for biopsy before and after receiving the test results and whether the 4Kscore test results influenced their decisions. Scores were groups into 3 risk categories: less than 7.5% ow risk; 7.5% to 19.9% intermediate risk; and 20% or more high risk. The physicians reported that the 4Kscore results influenced decisions in 89% of men and that the test led to a 64.4% reduction in prostate biopsies. The 4Kscore risk categories correlated highly (p<0.001) with biopsy outcomes in 171 men with biopsy results.
Absent direct evidence of clinical utility, a chain of evidence might be constructed. The 4Kscore test is associated with a diagnosis of aggressive prostate cancer. The incremental value of the 4Kscore concerning clinical examination and risk calculators in the intended use population is unknown due to deficiencies in estimating clinical validity described above. There is no prospective evidence that use of 4Kscore changes management decisions. The chain of evidence is incomplete.
At least 13 studies have reported on clinical validity of the KLK biomarkers, but only three used the marketed version of the 4Kscore test. The eligibility criteria for these studies had a lower limit for screening PSA but no upper limit. Given that the test manufacturer’s website states that the test is for men with inconclusive results, the inclusion of men with PSA levels greater than 10 ng/mL and positive DRE in the validation studies is likely not reflective of the intended use population. Studies that provide data on the incremental value of the components of the test show only small improvements with the intact PSA and hKA components (components specific to the 4Kscore). The studies performed in the U.S. men did not provide estimates (with confidence intervals) of validity compared with a standard clinical examination with a ratio of free or unbound PSA to total PSA (percent free PSA). Very few data are available on longer term clinical outcomes of men who were not biopsied base don 4Kscore results. No direct evidence supports the clinical utility of the test, and the chain of evidence is incomplete due to the limitations in estimates of clinical validity and utility.
The Prostate Health Index (phi) (Beckman Coulter) is an assay combining results of 3 blood serum immunoassays (tPSA (total PSA), fPSA (free PSA) and proPSA (p2PSA)) numerically to produce a “phi score”. This score is calculated in a routine laboratory using Beckman Coulter equipment and software with phi algorithm incorporated in the software. It has been suggested that the PSA isoform p2PSA might better distinguish between prostate cancer and benign prostatic conditions.
The phi score has been approved by FDA for distinguishing prostate cancer from benign prostatic conditions in men 50 years and older with above normal total PSA (tPSA) readings between 4.0 and 10 ng/mL who have had a negative DRE. The manufacturer’s website states that the test gives men “accurate information on what an elevated PSA level might mean and the probability of finding cancer on biopsy” and when “combined with family and patient history, the phi results can be used to determine the best individualized patient management decisions”.
Several systematic reviews and meta-analyses have described the clinical validity of p2PSA and phi. Results of systematic reviews and meta-analyses show most of the primary studies as low quality due to the design (most were retrospective), lack of blinding of outcome assessors to reference standard results, lack of clear cutoffs for diagnosis, and lack of explicit diagnostic question. Pooled estimates had high heterogeneity across studies but with generally low specificity of phi at 90% sensitivity.
The pivotal study described in the FDA SSED (summary of safety and effectiveness data) included men 50 years and older with nonsuspicious DRE and PSA levels between 4 ng/mL and 10 ng/mL who had a histologically confirmed diagnosis. The null hypothesis was that the phi specificity at 95% sensitivity would be no greater than the specificity of percent free PSA. Seven sites in the United States enrolled 658 men between 2008 and 2009 (97% enrolled prospectively, 3% enrolled retrospectively). Eighty-one percent of participants were white, 5% were African American, and 1% was Asian. At 95% sensitivity, using a phi cutoff of 22.1, the specificity was 14.1% (precision not reported) for phi compared with 9.9% for percent free PSA. AUC was 0.71 (95% CI, 0.67 to 0.75) for phi compared with 0.65 (95% CI, 0.61 to 0.69) for percent free PSA.
Additional studies have been published since the systematic reviews. In 2015 Fossati et. al. conducted a case-control study with 1036 European men younger than 60 years of age. They reported that phi had a higher AUC (0.70; 95% CI, 0.64 to 0.76) than total PSA (0.55; 95% CI, 0.48 to 0.61) in men younger than 60 years for detecting any prostate cancer. At 91% sensitivity, phi and total PSA had similar specificity (11.1% [95% CI, 6.8% to 16.8%] vs 10.5% [95% CI, 6.4% to 16.1%]) and an NPV (76.0% [95% CI, 59.3% to 92.7%] vs 75.0% [95% CI, 57.7% to 92.3%]), respectively. At the best combination of sensitivity and specificity (phi cutoff ≥41.2, total PSA cutoff ≥5.72), phi had a sensitivity of 64.2% (95% CI, 51.5% to 75.5%), a specificity of 63.2% (95% CI, 55.5% to 70.4%), and an NPV of 81.8% (95% CI, 75.2% to 88.4%) while total PSA had a sensitivity of 52.2% (95% CI, 39.7% to 64.6%), a specificity of 52.0% (95% CI, 44.3% to 59.7%), and an NPV of 73.6% (95% CI, 65.7% to 81.4%). A decision curve analysis found that using a model with age, prostate volume, total PSA, free PSA, percent free PSA, and phi with a probability cutoff of 10% would avoid 13% of biopsies while missing 0% of cancers; a cutoff of 20% would avoid 51% of biopsies while missing 18% of cancers; and a cutoff of 50% would avoid 94% of biopsies while missing 66% of cancers.
In 2016 Boegemann et. al. reported on results of a study of 769 European men ages 65 years and younger scheduled for initial or repeat prostate biopsy who were prospectively and retrospectively enrolled. The investigators compared phi with other PSA measures for detecting clinically significant vs insignificant cancer (PRIAS-criteria: T-stage T1c/T2; Gleason score ≤6; number of positive cores per biopsies ≤2; total PSA ≤10 ng/mL; PSA density <0.2 ng/mL). The AUC for phi (0.72; 95% CI, 0.68 to 0.76) was higher than that for total PSA (0.62; 95% CI, 0.58 to 0.66) or percent free PSA (0.64; 95% CI, 0.60 to 0.68).
Morote et. al. (2016) reported numerically higher but not statistically significantly higher area under the curve (AUC) for phi compared with total PSA or total free PSA for detecting aggressive prostate cancer in 357 men with PSA levels between 3 ng/mL and 10 ng/mL scheduled for first biopsy in a retrospective study in Spain. Similarly, Yu et al (2016) reported numerically but not statistically significantly higher AUC for phi vs total PSA in 114 men in China with PSA levels between 2 ng/mL and 10 ng/mL and negative DRE.
Porpiglia et. al. (2016) reported on the results of an observational retrospective study of 120 men with prostate cancer who received radical prostatectomy but were eligible for active surveillance. Multi-parametric MRI (mpMRI), phi, and PCA3 were performed on the single cohort. The base model for predicting pathologically confirmed significant prostate cancer (PCSPCa) had an area under the curve (AUC) of 0.71. Relative to the base model for predicting PCSPCa, phi increased the AUC by 4% (0.75; p<0.01), PCA3 increased the AUC by 1% (0.72; p<0.01), and mpMRI increased the AUC by 7% (0.78; p<0.01). Results of mpMRI provided the greatest net benefit in predicting the presence of PCSPCa while phi provided a small incremental benefit in prediction.
Many studies and systematic reviews of these studies have reported on the clinical validity of phi. In general, the comparator was a component of PSA (total PSA, free PSA) but have not included other risk factors from a standard clinical exam. Most of the primary studies included men with positive, negative, and inconclusive DRE and men with PSA levels outside of the 4- to 10-ng/mL range. African Americans have a high burden of morbidity and mortality but were not well represented in the study populations. There is no standardization of cutoffs used in a clinical setting for diagnosis and data on the diagnostic accuracy of phi for distinguishing clinically significant from insignificant cancer are lacking. A direct comparison between phi and mpMRI demonstrated a higher net benefit for mpMRI in predicting prostate cancer.
No studies directly measuring the effect of phi on clinical outcomes were found. A chain of evidence might be used to demonstrate clinical utility if each link in the chain is intact. The phi test is associated with a diagnosis of prostate cancer, although data on association with a diagnosis of aggressive prostate cancer are lacking. The phi test provided better diagnostic information than other measures of PSA alone, but decisions made with phi result plus other risk factors from clinical examination were not provided in most studies. Optimal cutoffs for classifying men into risk groups have not been standardized. No studies were found describing differences in management based on phi risk assessment. The chain of evidence is incomplete.
Systemic reviews including a NICE (National Institute for Health and Clinical Excellence) assesment have been reported and included many primary studies. In general, selected studies included some men outside of the intended use population (PSA levels outside of the 4 to 10 ng/mL range and abnormal DRE). Comparisons to diagnosis with clinical examination are lacking. The cutoffs for categorizing men into risk groups in clinical practice have not been standardized and therefore there is heterogeneity in reporting of performance characteristics and decision curve analysis. No studies were found describing differences in management based on phi risk assessment.
APIFINY technology is based on the measurement of eight prostate cancer specific biomarkers (autoantibodies) ARF 6, NKX3-1, 5-UTR-BMI1, CEP 164, 3-UTR-Ropporin, Desmocollin, AURKAIP-1, CSNK2A2. These biomarkers (autoantibodies) are produced and replicated (amplified) by the immune system in response to the presence of prostate cancer cells. The biomarkers (autoantibodies) are stable and, because of their amplifications are likely to be abundant and easy to detect, especially during the early stages of cancer.
Given the complexity of cancer risk assessment, obtaining additional information may provide insight to better inform an important clinical decisions such as an initial or repeat biopsy:
Statistical analysis shows there is an interdependence among the biomarkers (autoantibodies). Three of the biomarkers are associated with androgen-response regulation, and four are related to cellular structural integrity. The eighth biomarker has been implicated in prostate cancer progression and a variety of cellular functions ranging from cellular signaling for numerous protein kinases to regulating cell cycle and cell division. The APIFINY test process is performed in part using a qualitative immunoassay technique and in part using flow cytometry. The laboratory data generated by these methodologies are then subjected to a proprietary algorithmic analysis that generates a cancer risk score. APIFINY score reporting was designed to optimize the identification of patients at lower risk. Patients with a lower risk APIFINY score may be placed on a routine clinical monitoring program (i.e. semi-annual or annual check-up) with other accepted methods to assess the ongoing risk of prostate cancer. Geno typical men with higher APIFINY scores may require a more specific risk-assessment plan, which may include biopsy. Scores below 59 are considered lower relative risk, scores at or above 59 have a higher relative risk of prostate cancer.
Two studies have been done, a biomarker selection/algorithm development study and a clinical validation study. 519 samples were used in the biomarker selection/algorithm development study and 259 different samples were used in the clinical validation study. Although the studies are promising research has not yet been completed in determining the effects of age, race or other factors on the APIFINY score. Further studies are needed to determine the effects of demographics such as age, race or other factors on the APIFINY score and for clinical utility. Clinical utility of APIFINY test is uncertain, currently there is no evidence that the use of APIFINY tests can change management in ways that improve outcomes. The evidence is insufficient to determine the effects of this technology on net health outcomes.
ExoDx Prostate (IntelliScore), also called EPI, is a non-digital rectal exam (DRE) urine based liquid biopsy test that predicts the presence of high grade (Gleason score ≥ 7) prostate cancer for men 50 years of age and older with a PSA 2-10 ng/mL presenting for an initial biopsy. A “rule out” test ExoDx Prostate (IntelliScore) is designed to more accurately predict whether a patient presenting for an initial biopsy does not have a high grade prostate cancer and, thus could potentially avoid an initial biopsy and instead continue to be monitored. Using a proprietary algorithm that combines the relative weighted expression of the three gene signature, the test assigns an individual risk score for patients ranging from 0 to 100. A score > 15.6 is associated with an increased likelihood of high grade prostate cancer on biopsy. Physicians can utilize the score in conjunction with the other standard of care prognostic information to determine whether to proceed with a tissue biopsy.
McKiernan et. al. (2016) studied the performance of novel urine exosome gene expression assay (the ExoDx Prostate IntelliScore urine exosome assay) plus standard of care (SOC) (i.e. prostate specific antigen (PSA) level, age, race and family history) versus SOC alone for discriminating between Gleason score 7 and 6 and benign disease on initial biopsy. In training, using reverse transcriptase polymerase chain reaction (PCR), they compared the urine exosome gene expression assay with biopsy outcomes in 499 patients with prostate-specific antigen (PSA) level of 2 to 20 ng/mL. The derived prognostic score was then validated in 1064 patients from 22 community practice and academic urology clinical sites in the United States. Eligible participants included PCA free men, 50 years or older, scheduled for an initial or repeated prostate needle biopsy due to suspicious digital rectal examination (DRE) findings and/or PSA levels (limit range 2.0-20.0 ng/mL). In 255 men in the training target population (median age 62 years and median PSA level 5.0 ng/mL, and initial biopsy), the urine exosome gene expression assay plus SOC was associated with improved discrimination between Gleason score 7 or greater and Gleason score 6 and benign disease. Area under the curve (AUC) 0.77 (95% CI, 0.71-0.83) versus SOC ACU 0.66 (95% CI, 0.58-0.72) (P<.001) Independent validation in 519 patients urine exosome gene expression assay plus SOC AUC 0.73 (95% CI, 0.68-0.77) compared to SOC AUC 0.63 (95% CI, 0.58-0.68) (P<.001). Using a predefined cut point, 138 of 519 (27%) biopsies would have been avoided, missing only 5% of patients with dominant pattern for high risk Gleason score 7 disease. Limitations of this study include the DRE and free PSA as part of the standard of care variables. The limited accuracy of the DRE and the observed AUC’s for blood based assays that incorporate free-PSA suggests that the absence of these variables should not have a detrimental impact on overall performance of the exosome assay. Another limitation is that a central pathology review was not used, however, the objective was to evaluate the assay in a broad academic and community practice setting where individual pathology networks are the acceptable standard. The authors concluded the urine exosome gene expression assay was associated with improved identification of patients with higher grade prostate cancer among men with elevated PSA levels and could reduce the total number of unnecessary biopsies. Future efforts will compare the exosome test with some of the currently available blood based assays (when feasible), assess the impact of advanced imaging studies, which include MRI targeted biopsy assessment, and evaluate performance with respect to the pathologic abnormalities in the prostatectomy specimen. In addition, need to explore the role of the ExoDx Prostate IntelliScore in men enrolled in active surveillance protocols.
NCCN Guideline Version 2.2018 Prostate Early Detection, includes the following regarding this test: Given the lack of validation of the models/algorithms in independent publications, the unclear behavior in other screened populations, and the lack of clarity regarding the incremental value and cost effectiveness of this assay, the panel cannot recommended the routine use of this test at this time. Longer term follow-up of the cohorts to determine whether missed prostate cancers were ultimately detected is needed. In addition, validation of these tests in other cohorts of men is needed before they can be accepted as alternatives to (or perhaps preferable to) other tests. The NCCN Panel considers ExoDx Prostate (IntelliScore), also called EPI to be investigational at the present time.
There is no direct evidence that supports the clinical utility of the test, and the chain of evidence is incomplete due to the limitations in clinical validity and utility.
The Mi-Prostate score (MiPS) assay measures total serum PSA and post-DRE urine expression of the T2-ERG fusion (TMPRSS2: ERG) as well as another marker, PCA3. The test also predicts risk for having an aggressive tumor, helping doctors and patients make decisions about whether to wait and monitor test levels or pursue immediate biopsy.
In a validation study Sanda et. al. (2017) evaluated the priori primary hypothesis that combined measurement of PCA3 and TMPRSS2:ERG (T2;ERG) RNA in the urine after digital rectal exam (DRE) would improve specificity over measurement of prostate specific antigen alone for detecting cancer with Gleason score of 7 or higher. As a secondary objective, to evaluate the potential effect of such urine RNA testing on health care costs. Prospective, multicenter diagnostic evaluation and validation in academic and community based ambulatory urology clinics. Participants were a referred sample of men presenting for first-time prostate biopsy without pre-existing prostate cancer: 516 eligible participants from among 748 cohort participants in the developmental cohort and 561 eligible participants from 928 in the validation cohort. Urinary PCA3 and T2:ERG RNA measurements were taken before the prostate biopsy. The main outcome and measures, presence of prostate cancer having a Gleason score 7 or higher on prostate biopsy. Pathology testing was blinded to urine assay results. In the developmental cohort, a multiplex decision algorithm was constructed using urine RNA assays to optimize specificity while maintaining 95% sensitivity for predicting aggressive prostate cancer at initial biopsy. Findings were validated in a separate multicenter cohort via prespecified analysis, blinded per prospective-specimen-collection, retrospective-blinded-evaluation (PRoBE) criteria. Cost effects of the urinary testing strategy were evaluated by modeling observed biopsy results and previously reported treatment outcomes. Among the 516 men in the developmental cohort (mean age, 62 years; range, 33-85 years) combining testing of urinary T2:ERG and PCA3 at thresholds that preserved 95% sensitivity for detecting aggressive prostate cancer improved specificity from 18% to 39%. Among the 561 men in the validation cohort (mean age, 62 years; range, 27-86 years), analysis confirmed improvement in specificity (from 17% to 33%; lower bound of 1-sided 95% CI, 0.73%; prespecified 1-sided P = .04), while high sensitivity (93%) was preserved for aggressive prostate cancer detection. Forty-two percent of unnecessary prostate biopsies would have been averted by using the urine assay results to select men for biopsy. Cost analysis suggested that this urinary testing algorithm to restrict prostate biopsy has greater potential cost-benefit in younger men.
NCCN Guideline Version 2.2018 Prostate Early Detection, includes the following regarding this test: Given the lack of validation of the models/algorithms in independent publications, the unclear behavior in other screened populations, and the lack of clarity regarding the incremental value and cost effectiveness of this assay, the panel cannot recommended the routine use of this test at this time. Longer term follow-up of the cohorts to determine whether missed prostate cancers were ultimately detected is needed. In addition, validation of these tests in other cohorts of men is needed before they can be accepted as alternatives to (or perhaps preferable to) other tests. The NCCN Panel considers Mi-Prostate score (MiPS) to be investigational at the present time.
There is no direct evidence that supports the clinical utility of the test, and the chain of evidence is incomplete due to the limitations in clinical validity and utility.
SelectMDx helps identify patients at increased risk for aggressive disease, thereby aiding in the selection of men for prostate biopsy. SelectMDx for Prostate Cancer is a reverse transcription PCR (RT-PCR) assay performed on post-DRE (digital rectal examination), first void urine specimen from patients with clinical risk factors for prostate cancer, who are being considered for biopsy. The test measures the mRNA levels of the DLX1 and HOXC6 biomarkers, using KLK3 expression as internal reference, to aid in patient selection for prostate biopsy. Higher expression levels of DLX1 and HOXC6 mRNA are associated with an increased probability for high grade (Gleason Score (GS) ≥ 7) prostate cancer.
The assay was developed on an initial training set of 519 patients from 2 prospective multicenter studies and was then validated in a separate set of 386 patients from these trials. Using the expression of DLX1 and HOXC6 alone resulted in an AUC of 0.76, a sensitivity of 91%, a specificity of 36%, an NPV of 94%, and a PPV of 27% for the prediction of Gleason score ≥ 7 prostate cancer. When combined with PSA levels, PSAD, DRE results, age and family history in a multimodal model, the overall area under the curve (AUC) was 0.90 in the training set and 0.86 (95% CI, 0.80-0.92) in the validation set. A retrospective observational study compared results of SelectMDx with mpMRI (multi-parametric MRI) results in 172 patients who had mpMRI because of persistent clinical suspicion of prostate cancer or for local staging after positive biopsy. The AUC of SelectMDx for the prediction of mpMRI outcome was 0.83, whereas the AUC for PSA and PCA2 were 0.66 and 0.65, respectively.
NCCN Guideline Version 2.2018 Prostate Early Detection, includes the following regarding this test: Given the lack of validation of the models/algorithms in independent publications, the unclear behavior in other screened populations, and the lack of clarity regarding the incremental value and cost effectiveness of this assay, the panel cannot recommended the routine use of this test at this time. Longer term follow-up of the cohorts to determine whether missed prostate cancers were ultimately detected is needed. In addition, validation of these tests in other cohorts of men is needed before they can be accepted as alternatives to (or perhaps preferable to) other tests. The NCCN Panel considers SelectMDx to be investigational at the present time.
There is no direct evidence that supports the clinical utility of the test, and the chain of evidence is incomplete due to the limitations in clinical validity and utility.
The purpose of genetic and protein biomarker testing for prostate cancer is to inform the selection of men who should undergo repeat biopsy. The conventional decision-making tools for identifying men for prostate biopsy include DRE, serum PSA, and patient risk factors such as age, race, and family history of prostate cancer are previously described in the section for selection of men for initial prostate biopsy.
Given the risk, discomfort, burden of biopsy, and the low diagnostic yield, there is a need for noninvasive tests that distinguish potentially aggressive tumors that should be referred for rebiopsy from clinically insignificant localized tumors or other prostatic conditions that do not need rebiopsy with the goal of avoiding low-yield biopsy.
The relevant populations are men for whom a rebiopsy is being considered because the results of an initial prostate biopsy were negative or equivocal and other clinical symptoms remain suspicious.
Standard clinical examination for determining who goes to biopsy might include DRE, review of history of PSA values, along with consideration of risk factors such as age, race, and family history. The ratio of free (unbound) PSA to total PSA is lower in men who have prostate cancer than in those who do not. A percent free PSA (%fPSA) cutoff of 25% has been shown to have a sensitivity and specificity of 95% and 20%, respectively, for a group of men with total PSA levels between 4.0 ng/mL and 10.0 ng/mL.
The best way to combine all of the risk information to determine who should go to biopsy is not standardized. Risk algorithms have been developed that incorporate clinical risk factors into a risk score or probability. Two examples are the Prostate Cancer Prevention Trial (PCPT) predictive model and the Rotterdam Prostate Cancer risk calculator (also known as the European Research Screening Prostate Cancer Risk Calculator 4 (ERSPC-RC). The AUA and the Society of Abdominal Radiology recently recommended that high-quality prostate MRI, if available, should be strongly considered in any patient with a prior negative biopsy who has persistent clinical suspicion for prostate cancer and who is under evaluation for a possible repeat biopsy.
The beneficial outcome of the test is to avoid a negative biopsy for prostate cancer. A harmful outcome is failure to undergo a biopsy that would be positive for prostate cancer, especially when disease is advanced or aggressive. Thus the relevant measures of clinical validity are the sensitivity and negative predictive value. The appropriate reference standard is biopsy, though prostate biopsy is an imperfect diagnostic tool. Biopsies can miss cancers and repeat biopsies are sometimes needed to confirm the diagnosis. Detection rates vary by biopsy method and patient characteristics, with published estimates between 10% and 28% for a second biopsy and 5% and 10% for a third biopsy.
The timeframe of interest for calculating performance characteristics is time to biopsy result. Men who forgo biopsy based on test results could miss or delay diagnosis of cancer. Longer follow-up would be necessary to determine effects on overall survival.
Screening using PSA levels and DRE may be performed in the primary care setting with referral to specialty (urologist) care for suspicious findings and biopsy. Clinical practice on screening methods and frequency vary widely.
PCA3 (prostate cancer gene 3) is overexpressed in prostate cancer, and PCA3 messenger (mRNA) can be detectied in urine samples following a digital rectal exam (DRE). When normalized using PSA to account for prostate cells released into the urine (PCA3 score), the test has significantly improved specificity compared with serum PSA and may better discriminate patients with benign findings on (first or second) biopsy from those with malignant biopsy results.
The Progensa PCA3 assay (Hologic Gen-Probe) has been approved by the FDA to aid in the decision for repeat biopsy in men 50 years or older who have had 1 or more negative prostate biopsies and for whom a repeat biopsy would be recommended based on current standard of care. The Progensa PCA3 assay should not be used for men with atypical small acinar proliferation on their most recent biopsy. The manufacturer's website states that the test is intended to identify men who have negative first biopsy results to determine who needs a follow-up biopsy and that a PCA3 score less than 25 is associated with a decreased likelihood of a positive biopsy.
Several systematic reviews and meta-analyses have described the clinical validity of Progensa PCA3. The majority of the studies were observational, although one study used the placebo arm from a randomized controlled trial and validation trial. Reviewers selected studies of men with positive, negative, or inconclusive DRE without restrictions on PSA levels.
Nicholson et. al. (2015) evaluated the clinical effectiveness and cost effectiveness of the PCA3 assay and the phi in the diagnosis of prostate cancer. The assessment of clinical effectiveness involved three separate systematic reviews, namely reviews of the analytical validity, the clinical validity and clinical utility of these tests. The assessment of cost-effectiveness comprised a systematic review of full economic evaluations and the development of a de novo economic model. This review included men suspected of having prostate cancer for whom the results of an initial prostate biopsy were negative or equivocal. The use of PCA3 score or phi in combination with existing tests (including histopathology results, prostate-specific antigen lvel and digital rectal examination), multiparametric MRI and clinical judgement. In addition to documents published by the manufacturers, six studies were identified for inclusion in the analytic validity review. The review identified issues concerning the precision of the PCA3 assay measurements. It also highlighted issues relating to the storage requirements and stability of samples intended for analysis using the phi assay. Fifteen studies met inclusion criteria for the clinical validity review. These studies reported results for 10 different clinical comparisons. There was insufficient evidence to enable the identification of appropriate test threshold values for use in a clinical setting. In addition, the implications of adding either the PCA3 assay or the phi to clinical assessment were not clear. Furthermore, the additional of the PCA assay or the phi to clinical assessment plus MRI imaging was not found to improve discrimination. No published papers met the inclusion criteria for the clinical utility review. The results from the cost effectiveness analyses indicated that using either the PCA3 or the phi was not cost effective. The authors concluded, the clinical benefit of using the PCA3 assay or the phi in combination with existing tests, scans and clinical judgement has not yet been confirmed. The results from the cost effectiveness analyses indicate the use of these tests would not be cost effective.
Cui et. al. (2016) investigated the clinical value of the urine PCA3 test in the diagnosis of prostate cancer by pooling the published data. A total of 46 clinical trials included 12,295 subjects were included in the meta-analysis. The pooled sensitivity, specificity, positive likelihood ratio (+LR), negative likelihood ratio (-LR), diagnostic odds ratio (DOR) and area under the curve (AUC) were 0.65 (95% confidence interval [CI]: 0.63-0.66), 0.73 (95% CI: 0.72-0.74), 2.23 (95% CI: 1.91-2.62), 0.48 (95% CI: 0.44-0.52), 5.31 (95% CI: 4.19-6.73) and 0.75 (95% CI: 0.74-0.77), respectively. This meta-analysis has some limitations. First, the study numbers and the heterogeneity of their approaches influenced the accuracy. Although the gold standard (biopsy) was used in all studies, the patient selection, lack of blinding, and different PCA3 cut-off values caused heterogeneity. The authors concluded, according to the current meta-analysis, the PCA3 test shows good diagnostic performance. However, it requires further exploration in well-designed and appropriately powered trials to determine intermediate and long-term outcomes. Long term observational studies of health outcomes are also subject to biases.
In 2014 the National Cancer Institute conducted a prospective validation trial to assess the diagnostic performance of the prostate cancer antigen 3 (PCA3) urinary assay for the detection of prostate cancer among men screened with PSA. The target population included men who had been screened for prostate cancer, primarily with a PSA test, some of whom had undergone a previous prostate biopsy. The study included 859 men (mean age, 62 years) from 11 centers in the United States scheduled for a diagnostic prostate biopsy between December 2009 and June 2011 were enrolled. The primary outcome was to assess whether PCA3 could improve the positive predictive value (PPV) for an initial biopsy (at a score > 60) and the negative predictive value (NPV) for a repeat biopsy (at a score < 20). For the detection of any cancer, PPV was 80% (95% CI, 72% to 86%) in the initial biopsy group, and NPV was 88% (95% CI, 81% to 93%) in the repeat biopsy group. The addition of PCA3 to individual risk estimation models (which included age, race/ethnicity, prior biopsy, PSA, and digital rectal examination) improved the stratification of cancer and of high-grade cancer. There were several limitations related to this study, the potential role of PCA3 as a screening test to replace PSA screening was not examined, and these findings do not extend to men who have not been prescreened. Patients were followed only through the index biopsy, and it is likely that some of the patients had false negative prostate biopsies, given the known sampling error of a needle biopsy for the detection of cancer and the under-detection of high grade prostate cancer.
The pivotal study describing in the FDA SSED (summary of safety and effectiveness data) for Progensa included 495 men from 15 clinical sites who had at least 1 negative prostate biopsy and whose urologist recommended repeat biopsy. Prostate biopsy was performed using each site’s local standard procedure. A total of 433 (87.5%) were white, 45 (9.1%) were African American, and 11 (2.2%) were Asian. A valid PCA3 score and biopsy result were available for 466 men. Using a PCA3 cutoff score of 25, the performance characteristics for positive biopsy were as follows: sensitivity, 77.5% (95% CI, 68.4% to 84.5%); specificity, 57.1% (95% CI, 52.0% to 62.1%); positive predictive value (PPV), 33.6% (95% CI, 30.0% to 37.2%); and negative predictive value (NPV), 90.0% (95% CI, 86.5% to 93.1%). In other words, 208 men in the study might have been spared an unnecessary repeat biopsy if a cutoff of 25 was used to recommend repeat biopsy. On the other hand, 23 of the men who had a biopsy positive for prostate cancer might have had their diagnosis delayed due to negative PCA3 result.
Clinical studies have compared PCA3 results with clinical examination and risk calculators; those focused on distinguishing between aggressive and indolent cancer are particularly relevant. Ankerst et. al. (2008) reported that incorporating the PCA3 score into the PCPT risk calculator improved the diagnostic accuracy of the calculator (from an area under the curve (AUC) of 0.653 to 0.696). Chun et. al. (2009), using a multivariate nomogram, demonstrated a 5% gain in predictive accuracy when PCA3 was incorporated with other predictive variables such as age, DRE results, PSA levels, prostate volume, and biopsy history. In a 2011 study of 218 patients with PSA levels of 10 ng/mL or less, Perdona et. al. performed a head-to-head comparison of these 2 risk assessment tools and suggested both might have value in clinical decision making.
Several studies have evaluated the PCA3 score as a tool for distinguishing between patients with indolent cancers who may need only active surveillance and those with aggressive cancers who warrant aggressive therapy. Three studies from 2008 Haese et. al., Nakanishi et. al., and Whitman et. al. demonstrated an association between PCA3 scores and evidence of tumor aggressiveness. However, these findings were not confirmed in a 2006 study by Bostwick et. al. or a 2008 study by vans Gils et. al. Auprich et. al. (2011) reported that PCA3 scores appeared to enhance identification of indolent disease but not pathologically advanced or aggressive cancer.
Tosoian et. al. (2010) reported on a short-term prospective cohort study evaluating the PCA3 score in relation to outcomes in an active surveillance program involving 294 patients. The PCA3 score did not distinguish between patients who had stable disease and those with more aggressive features.
Clinical utility studies using assay results for decision making for initial biopsy, repeat biopsy, or treatment have not been reported, nor have studies of the effects of using assay results on clinical outcomes. Several studies using decision analysis to estimate the cost-benefit tradeoff between reduction in unnecessary biopsies and missed prostate cancers have been published. One group reported potential reductions in unnecessary biopsies of 48% to 52%, with attendant increases in missed prostate cancers of 6% to 15% using either a PCA3-based nomogram or PCA3 level corrected for prostate volume (PCA3 density). Although both studies were prospective, neither assessed utility of the test for clinical decision making because all patients underwent biopsy. Merdan et. al. (2015) used decision analysis to simulate long-term outcomes associated with use of the PCA3 score to trigger repeat biopsy compared with the PCPT risk calculator in men with at least 1 previous negative biopsy and elevated PSA levels. They estimated that incorporating the PCA3 score of 25 (biopsy threshold) into the decision to recommend repeat biopsy could avoid 55.4% of repeat biopsies, with a 0.93% reduction in the 10-year survival rate.
Given the lack of direct evidence of utility, a chain of evidence would be needed to demonstrate clinical utility. The PCA3 test is associated with a diagnosis of prostate cancer, although data on its association with a diagnosis of aggressive prostate cancer are lacking. The PCA3 test provided better diagnostic information than other measures of PSA, but comparison with decisions made using risk factors from clinical examination was not provided in most studies. No prospective studies were found describing differences in management based on PCA3 risk assessment. The chain of evidence is incomplete.
Systematic reviews, including a health technology assessment, have been reported and included many primary studies. Studies of the PCA3 score as a diagnostic test for prostate cancer have reported sensitivities and specificities in the moderate range. In general, these studies are preliminary and report on clinical performance characteristics in different populations and with various assay cutoff values, reflecting the lack of standardization in performance and interpretation of PCA3 test results. Cutoffs for recommending repeat biopsy with the Progensa test have been suggested by the manufacturer and were used in a validation study for FDA approval. The clinical utility of the PCA3 test is uncertain because there is no evidence that its use can change management in ways that improve outcomes.
One of the epigenetic mechanisms that is considered to be involved in the development of prostate cancer is DNA methylation. Hypermethylation within promotor region of tumor suppressor genes is an important mechanism of gene inactivation and has been described for many different tumor types. These types of alterations are also potentially reversible, unlike genetic alterations such as mutations, which may lead them being considered as possible targets for gene therapy. Currently, aberrant promoter hypermethylation has been investigated in specific genes from the following groups: tumor-suppressor genes, proto-oncogenes, genes involved in cell adhesion, and genes involved in cell-cycle regulation. Glutathione S-transferase P1 (GSTP1) is the most widely studied methylation markers for prostate cancer, usually as a diagnostic application. Several studies reported associations between DNA hypermethylation at various gen loci (RASSF1A, APC, GSTP1, PTGS2, RQQR-beta, TIG1, AOX1, C1orf114, GAS6, HAPLN3, KLF8, MOB3B) and prostate cancer. It has been suggested that a valuable first step in diagnostic use might be to test for methylated genes to select patients undergoing prostate biopsy who might not require a repeat biopsy.
In a 2012 meta-analysis by Van Neste et. al., 30 peer-reviewed studies of hypermethylation of GSTP1 and other genes in prostate tissue were evaluated.88 The pooled estimates of sensitivity for GSTP1 to distinguish prostate cancer from normal in biopsies (328 cases, 263 controls) was 82%, with 95% specificity, 95% NPV (negative predictive value), and 85% PPV (positive predictive value). The combination of GSTP1, APC, and RARβ had a sensitivity of 95%, specificity of 95%, NPV of 99%, and PPV of 95%. Reviewers suggested that a valuable first step in diagnostic use might be to test for methylated genes to select patients undergoing prostate biopsy who might not require repeat biopsy.
Following the 2012 meta-analysis, several studies reported on associations between DNA hypermethylation at various gene loci (RASSF1A, APC, GSTP1, PTGS2, RARβ, TIG1, AOX1, C1orf114, GAS6, HAPLN3, KLF8, MOB3B) and prostate cancer. In contrast, Kachakova et. al (2013) found that HIST1H4K hypermethylation was more likely due to aging than to prostate carcinogenesis.
ConfirmMDx is a test for gene methylation intended to distinguish true from false negative prostate biopsies to avoid the need for repeat biopsy in cases of true negative and to identify geno typical men who may need a repeat biopsy. The test measures methylation of the genes GSTP1, APC and RASSF-1.
Two blinded multicenter validation studies of the ConfirmMDx test have been performed. Partin et. al. (2014) reported on results of the DOCUMENT study; it evaluated archived, cancer-negative prostate biopsy core tissue samples from 350 men from 5 U.S. urology centers. All patients underwent repeat biopsy within 24 months. Men with 2 consecutive negative biopsies were classified as controls and men with a negative biopsy followed by a positive biopsy were classified as cases. Thirty men (9%) were excluded from analysis because of non-eligibility (n=2), insufficient DNA (n=1), insufficient biopsy cores (n=23), or detection of adenocarcinoma in the first biopsy based on central pathology review (n=4); 320 men were included in analysis (92 cases, 228 controls). Median age was 62 years (range, not given). Median PSA level was 5.3 ng/mL; 23% of men had PSA levels less than 4 ng/mL and 10% had a PSA level of 10 ng/mL or higher. Sixty percent of men had a normal DRE. Forty-two (13%) of the men were black, 232 (73%) were white, and 13 (4%) were Asian. The ConfirmMDx test, performed on the first biopsy, resulted in a NPV (negative predictive value) of 88% (95% CI, 85% to 91%), sensitivity of 62% (95% CI, 51% to 72%), and specificity of 64% (95% CI, 57% to 70%). The study was not powered to determine accurately the performance characteristics in a subgroup of black patients, but the estimated sensitivity was 77% (95% CI, 46% to 95%), specificity was 66% (95% CI, 46% to 82%), and NPV was 93% (85% CI, 82% to 97%). Multivariate analysis of potential predictors of cancer on repeat biopsy, corrected for age, PSA, DRE, first biopsy histopathology characteristics, and race, showed that the ConfirmMDx test was the most significant independent predictor of patient outcome (OR=2.69; 95% CI, 1.60 to 4.51).
The MATLOC study, reported by Stewart et. al. (2013), tested archived cancer-negative prostate biopsy needle core tissue samples from 498 men from the U.K. and Belgium. Patients underwent repeat biopsy within 30 months; cases had a positive second biopsy while controls had a negative second biopsy. A total of 483 men were included in the analysis (87 cases, 396 controls). The median PSA level was 5.9 ng/mL; 21% of men had PSA levels less than 4 ng/mL and 18% had PSA levels of 10 ng/mL or higher. Seventy-three percent of men had benign DRE. The ConfirmMDx test, performed on the first biopsy, resulted in a NPV (negative predictive value) of 90% (95% CI, 87% to 93%), sensitivity of 68% (95% CI, 57% to 77%), and specificity of 64% (95% CI, 59% to 69%). Multivariate analysis of potential predictors of cancer on repeat biopsy, corrected for patient age, PSA, DRE, and first biopsy histopathology characteristics, showed that the ConfirmMDx test was the most significant independent predictor of patient outcome (OR=3.17; 95% CI, 1.81 to 5.53).
In 2016, Van Neste et. al. reported on results of combined data from the DOCUMENT and MATLOC studies to investigate whether DNA methylation intensities were associated with high-grade (Gleason score, ≥7) prostate cancer. DNA methylation was the most significant and important predictor of high-grade cancer, resulting in an NPV (negative predictive value) of 96% (precision not reported).
In 2014, Wojno et. al. reported on a field observation study in which practicing urologists at 5 centers used the ConfirmMDx test to evaluate at least 40 men with previous cancer-negative biopsies who were considered at risk for prostate cancer. Centers reported whether patients who had a negative test assay result had undergone a repeat biopsy at the time of the analysis. Median patient follow-up after the assay results were received was 9 months. A total of 138 patients were included in the analysis. The median PSA level was 4.7 ng/mL. Repeat biopsies had been performed in 6 (4.3%) of the 138 men with a negative ConfirmMDx test, in which no cancer was identified.
In 2013, Aubry et. al. analyzed the expected reduction in biopsies associated with ConfirmMDx use. Using the MATLOC estimates of performance characteristics for ConfirmMDx, the authors estimated that 1106 biopsies per 1 million people would be avoided. The study did not include a decision analysis comparing the tradeoff in reduction in biopsies and missed cancers.
MDxHealth completed enrollment into the PASCUAL trial in April 2015. The PASCUAL trial is described as an observational trial of ConfirmMDx to evaluate the impact of the test (clinical utility) on physician decisions for repeat biopsy. Results have not yet been published.
No studies were found that directly show the effects of using ConfirmMDx results on clinical outcomes. Given the lack of direct evidence of utility, a chain of evidence would be needed to demonstrate clinical utility. The ConfirmMDx test is associated with a diagnosis of prostate cancer and aggressive prostate cancer. The validity studies of the ConfirmMDx test included men in the intended use population but did not compare performance characteristics with clinical examination plus percent free PSA. One survey of urologists who had previously used the ConfirmMDx test found that most ConfirmMDx negative patients did not have a biopsy. Prospective data on utility should be available after completion of PASCUAL. No data are available on the longer term clinical outcomes of the men who did not have biopsy based on ConfirmMDx results. The chain of evidence is incomplete.
Two clinical validation studies have reported on the clinical validity of ConfirmMDx score in the intended use population. The studies did not provide estimates of validity compared to a standard clinical examination with percent free PSA (%fPSA). No data are available on the long term clinical outcomes or clinical utility of these tests. The indirect chain of evidence is incomplete due to the limitations in evidence on the comparative clinical validity and utility.
TMPRSS2 is an androgen-regulated transmembrane serine protease that is preferentially expressed in normal prostate tissue. In prostate cancer, it may be fused to an ETS (E26 transformation-specific) family transcription factor (ERG, ETV1, ETV4, or ETV5), which modulates transcription of target genes involved in cell growth, transformation and apoptosis. The result of gene fusion with an ETS transcription gene is that the androgen-responsive promoter of TMPRSS2 upregulates expression of the ETS gene, suggesting a mechanism for neoplastic transformation. Fusion genes may be detected in tissue, serum and urine.
TMPRSS2-ERG gene rearrangements have been reported in 50% or more of primary prostate cancer samples. Although ERG appears to be the most common ETS family transcription factor involved in the development of fusion genes, not all are associated with TIMPRSS2. About 6% of observed rearrangements are seen with SLC45A3, and about 5% appear to involve other types of rearrangement. Attention has been directed at using post-DRE urine samples to look for fusion genes as markers of prostate cancer.
In 2014, Yao et. al. published a systematic review with meta-analysis of TMPRSS2-ERG for the detection of prostate cancer. Literature was searched through July 2013, and 32 articles were included. Pooled sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio were 47% (95% CI, 46% to 49%), 93% (95% CI, 92% to 94%), 8.9 (95% CI, 5.7 to 14.1), and 0.49 (95% CI, 0.43 to 0.55), respectively. Statistical heterogeneity was high (I2>85%). It was unclear whether studies in screening populations were pooled with enriched patient samples (e.g., elevated PSA and/or negative biopsy). There also was variability in the type of tissue samples analyzed (urine, prostatic secretions, biopsy, surgical specimens); the type of TMPRSS2-ERG assays used (fluorescence in situ hybridization, immunohistochemistry, real-time reverse transcriptase PCR, transcription-mediated amplification); and in TMPRSS2-ERG threshold cutoff values.
As described above under testing for initial prostate biopsy, Mi-Prostate Score (MiPS) can also be utilized in individuals who had prior negative biopsy, to determine if repeat biopsy should be performed.
Tomlins et. al. (2011) developed a transcription-mediated amplification assay to measure TMPRSS2-ERG fusion transcripts in parallel with PCA3. Combining results from the TMPRSS2-ERG and PCA tests and incorporating them into the multivariate PCPT risk calculator appeared to improve identification of patients with clinically significant cancer using Epstein criteria and high-grade cancer on biopsy. Although the study was large (1312 men at multiple centers), it was confounded by assay modifications during the study and by use of cross-validation rather than independent validation, using independent training and testing sets.
In 2013, this same group evaluated 45 men using a multivariable algorithm that included serum PSA plus urine TMPSS2-ERG and PCA3 from a post-DRE sample. Samples were collected before prostate biopsy at 2 centers. For cancer prediction, sensitivity and specificity were 80% and 90%, respectively. Area under the curve (AUC) was 0.88.
In 2016, Tomlins et. al. published results of a validation study of the MiPS score in 1244 prospectively collected, post-DRE urine samples from 7 U.S. clinics. A total of 1225 of the specimens had sufficient materials for both TMPSS2-ERG and PCA3 analysis and were included. Eighty percent of patients were presenting for initial biopsy. Seventy-three percent were white; the percentage of African Americans was not given. Approximately 25% of the men were older than 70. Twenty-three percent had an abnormal DRE, and the median PSA level was 4.7 ng/mL. The AUCs for predicting any cancer using PSA alone, PCPT risk calculator alone, and the MiPS score alone were 0.59, 0.64, and 0.76, respectively (CIs not given, p<0.001 for MiPS vs PCPT). The area under the curve (AUCs) for predicting high-grade cancer were 0.65, 0.71, and 0.78, respectively (p<0.001 for MiPS vs PCPT). A MiPS score threshold for recommending biopsy has not been provided, and so sensitivity and NPV (negative predictive value) were not calculated.
Tomlins et. al. (2016) also used decision-curve analysis to estimate the number of biopsies that would have been performed and cancers that would have been missed using a MiPS risk cutoff for biopsy in their cohort. Compared with a biopsy-all strategy, using a MiPS cutoff for aggressive cancer of 15% would have avoided 36% of biopsies while missing 7.0% of any prostate cancer and 1.6% of high-grade prostate cancer diagnoses. Using the PCPT risk calculator cutoff of 15% for aggressive cancer would have avoided 68% of biopsies while missing 25% of any cancer and 8% of high-grade cancer.
No studies were found that directly show the effects of using MiPS results on clinical outcomes. Given the lack of direct evidence of utility, a chain of evidence would be needed to demonstrate clinical utility. The MiPS test is associated with a diagnosis of prostate cancer and aggressive prostate cancer. The clinical validity study of the MiPS test included men with relevant PSA levels but also included men with positive DRE who would not likely forgo biopsy. The clinical validation study included comparison of performance characteristics with standard risk calculators; comparison with percent free PSA was not provided. Confirmation of performance characteristics is needed. No prospective data are available on using the MiPS score for decision making. No data are available on the longer term clinical outcomes of the men who did not have biopsy based on MiPS results. The chain is incomplete.
Concomitant detection of TMPRSS2:ERG and PCA3 may more accurately identify geno typical men with prostate cancer. However, current evidence is insufficient to support its use. Estimated accuracy varies across available studies. The Mi-Prostate Score (MiPS) test has preliminary data suggesting improved clinical validity compared to the PCPT risk calculator in a validation study but independent confirmation of clinical validity and comparison with percent free PSA (%fPSA) is needed. Data on clinical utility is lacking.
The Prostate Core Mitomics Test (PCMT; Mitomics; formerly Genesis Genomics) is a proprietary test that is intended to determine whether a patient has prostate cancer, despite a negative prostate biopsy, by analyzing deletions in the mitochondrial DNA by polymerase chain reaction (PCR) to detect “tumor field effect.” The test is performed on the initial negative prostate biopsy tissue. According to the company website, a negative PCMT result confirms the results of the negative biopsy (i.e. the patient does not have prostate cancer) and the patient can avoid a second biopsy, but a positive PCMT means the patient is at high risk of undiagnosed prostate cancer. The website also states that physicians should consider using PCMT for patients who have a negative initial biopsy but continue to have elevated PSA, rising PSA, irregular DRE, atypical small acinar proliferation, high-grade prostatic intraepithelial neoplasia or inconclusive biopsy.
A 2006 study retrospectively analyzed mitochondrial DNA variants from 3 tissue types from 24 prostatectomy specimens: prostate cancer, adjacent benign tissue, and benign tissue distant to the tumor (defined as tissue from a non-diseased lobe or at least 10-cell diameters from the tumor if in the same lobe). Prostate needle biopsy tissue (from 12 individuals referred for biopsy) that were histologically benign were used as controls. Results from the prostatectomy tissue analysis showed that 16 (66.7%) of 24 had variants in all 3 tissue types, 22 (91.7%) of 24 had variants in malignant samples, 19 (79.2%) of 24 in adjacent benign samples, and 22 of 24 in distant benign glands. Overall, 273 somatic variants were observed in this sample set. In the control group, 7 (58.3%) patients had between 1 and 5 genetic alterations, mainly in noncoding regions. The authors concluded that the variants found in the malignant group vs the control group differed significantly and that mitochondrial DNA variants are an indicator of malignant transformation in prostate tissue.
In 2008, Maki et.al. reported on the discovery and characterization of a 3.4-kilobase mitochondrial genome deletion and its association with prostate cancer. A pilot study analyzed 38 benign biopsy specimens from 22 patients, 41 malignant biopsy specimens from 24 patients, and 29 proximal to malignant (PTM) biopsy specimens from 22 patients. All patients with malignant biopsies had a subsequent prostatectomy, and the diagnosis of cancer was confirmed. The PTM biopsy samples were negative for cancer and were from the cohort that underwent prostatectomy. A confirmation study used 98 benign biopsy specimens from 22 patients, 75 malignant biopsy specimens from 65 patients, and 123 PTM biopsy specimens from 96 patients. In the confirmation study, patients had to have at least 2 successive negative biopsies; the first negative biopsy was used for analyses. For both the pilot and confirmation studies, samples for analysis were selected based on a review of pathology reports. The levels of the variation were measured by quantitative PCR and, based on PCR cycle threshold, data were used to calculate a score for each biopsy sample. In the pilot study, the scores were statistically significant between benign and malignant (p<0.000) and benign and proximal (p<0.003) samples. The PTM samples closely resembled the malignant sample, with no statistically significant resolution between the scores (p<0.833), to which the authors attributed to a field cancerization phenomenon. Results from the larger confirmation study were similar. Compared with histopathologic examination of the benign and malignant samples, the sensitivity and specificity were 80% and 71%, respectively, and the area under the receiver operating characteristics (AUROC) was 0.83 for the deletion. A blinded, external validation study showed a sensitivity and specificity of 83% and 79% and the AUROC of 0.87.
In 2010, Robinson et. al. assessed the clinical value of the 3.4 kilobase deletion described in the Maki study in predicting rebiopsy outcomes. Levels of the deletion were measured by quantitative PCR in prostate biopsies negative for cancer from 101 patients who underwent repeat biopsy within 1 year and had known outcomes. Of the 101 first biopsies, the diagnosis was normal in 8, atypical and/or had prostatic intraepithelial neoplasia in 50, and hyperplasia or inflammation in 43. The clinical performance of the deletion was calculated with the use of an empirically established cycle threshold cutoff, the lowest cycle threshold as diagnostic of prostate cancer, and the histopathologic diagnosis on second biopsy. Final data were based on 94 patients, who on second biopsy had 20 malignant and 74 benign diagnoses.
The cycle cutoff gave a sensitivity and specificity of 84% and 54%, respectively, with an AUROC of 0.75. NPV was 91%.
No peer-reviewed, full-length publications on the clinical utility of the commercially available Prostate Core Mitomics Test (PCMT) was identified.
The Prostate Core Mitomics Test (PCMT) has preliminary data on performance characteristics in small validation study but independent confirmation of clinical validity is needed. The studies did not provide estimates of validity compared to a standard clinical examination. No data is available on long term clinical outcomes. Data on clinical utility is lacking.
Since no single gene markers have been found that are both highly sensitive and highly specific for diagnosing prostate cancer, particularly in geno typical men already known to have an elevated PSA level, some investigators are combining several promising markers into a single diagnostic panel. Although promising in concept, only single studies of various panels have been published, and none apparently are offered as a clinical service.
Single nucleotide Variants (SNVs) occur when a single nucleotide is replaced with another, and they are the most common type of genetic variation in humans. They occur normally throughout the genome and can act as biological markers for disease association. Genome-wide association studies have identified associations between prostate cancer risk and specific SNVs. However, it is widely accepted that individually, SNV-associated disease risk is low and of no value in screening for disease, although multiple SNVs in combination may account for a higher proportion of prostate cancer. Investigators have begun to explore the use of algorithms incorporating information from multiple SNVs to increase the clinical value of testing.
Ma et. al. (2014) examined various algorithms for cancer diagnosis and prognosis using urine and plasma levels of multiple genes, including PCA3, PSA, TMPRSS2, and ERG. One algorithm distinguished prostate cancer from benign prostatic hypertrophy with an AUC of 0.78. Another algorithm distinguished men with a Gleason score 7 or higher for men with a Gleason score less than 7 (AUC=0.88). Combination of these 2 algorithms into a scoring system predicted the presence of a Gleason score 7 or higher in 75% of men. Qu et al (2013) reported on preliminary results of a 3-gene panel (androgen receptor [AR], PTEN, TMPRSS2-ERG) analyzed by fluorescence in situ hybridization.112 Thirty-one percent of 110 archived primary tumor samples and 97 metastatic tumor samples from a separate cohort of patients were analyzable. Chromosomal abnormalities were detected in 53% of primary prostate cancers and in 87% of metastatic tumors (p<0.001).
In 2015, Leyten et. al. reported on the development of a gene panel using specimens from 133 patients that included 3 urinary biomarkers (HOXC6, TDRD1, DLX1). When the gene panel was used with PSA, the combined AUC for predicting high-grade prostate cancer was 0.81 (95% CI, 0.75 to 0.86), which was higher than the concurrently measured Progensa AUC of 0.68 (95% CI, 0.62 to 0.75). Xiao et al (2016) reported the development of an 8-gene panel (PMP22, HPN, LMTK2, FN1, EZH2, GOLM1, PCA3, GSTP1) that was able to distinguish high-grade prostate cancer from indolent prostate cancer with a sensitivity of 93% (95% CI, 88% to 97%), a specificity of 70% (95% CI, 36% to 104%), a PPV of 98% (95% CI, 95% to 100%), and an NPV of 61% (95% CI, 25% to 97%) in a specimen cohort of 158 men.
A 2012 Agency for Healthcare Research and Quality report on multigene panels in prostate cancer risk assessment reviewed the literature on SNV panel tests for assessing risk of prostate cancer. All of the studies included in the review had poor discriminative ability for predicting risk of prostate cancer, had moderate risk of bias, and none of the panels had been evaluated in routine clinical settings. The conclusions of the review were that the evidence on currently available SNV panels does not permit meaningful assessment of analytic validity, the limited evidence on clinical validity is insufficient to conclude that SNV panels would perform adequately as a screening test and that there is no evidence available on the clinical utility of current panels.
Kader et. al. (2012) evaluated a panel of 33 prostate cancerâ€’associated SNVs identified from genome-wide association studies in 1654 men. Genetic score was a significant (p<0.001) independent predictor of prostate cancer (OR=1.72; 95% CI, 1.44 to 2.09) after adjusting for clinical variables and family history. Addition of genetic markers to the classification of prostate cancer risk resulted in 33% of men by reclassified into a different risk quartile. Approximately half of these (n=267) were downgraded to a lower risk quartile, and the other half (n=265) were upgraded into a higher risk quartile. The net reclassification benefit was 10% (p=0.002). The authors concluded that, with the additional information of genetic score, the same number of cancers could be detected but with 15% fewer biopsies.
In a 2010 review by Ioannidis et al, 27 gene variants across a large range of chromosomal locations were identified that increased risk for prostate cancer, although, in all cases, the observed incremental risk was modest (OR≤1.36).
Lindstrom et. al. (2011), in a study of 10,501 cases of prostate cancer and 10,831 controls, identified 36 SNVs showing association with prostate cancer risk, including two (rs2735893, rs266849) that showed differential association with Gleason score. Per allele odds ranged from 1.07 to 1.44.
Ishak and Giri (2011) reviewed 11 replication studies involving 30 SNVs (19 in men of African descent, 10 in men with familial prostate cancer). Odds ratios were positively associated with prostate cancer, although the magnitude of association was small (range, 1.11-2.63).
Numerous studies have demonstrated the association of many gene panels and SNVs with prostate cancer. These studies, in early stages of development, have generally shown a modest degree of association with future risk for prostate cancer. The clinical utility of these tests is uncertain; there is no evidence that information obtained from gene panels or SNV testing can be used to change management in ways that improve outcomes.
For individuals who are being considered for an initial prostate biopsy or a repeat biopsy who receive testing for genetic and protein biomarkers of prostate cancer, the evidence systematic reviews and meta-analyses and primarily observational studies. The evidence supporting clinical utility varies by test but has not been directly shown for any genetic or protein biomarker test. In general, the performance of genetic and protein biomarker testing for predicting biopsy results compared with clinical examination, including the ratio of free or unbound PSA to total PSA, is lacking. Procedures for referrals for biopsy based on clinical examination vary making it difficult to quantify performance characteristics for this comparator. There is considerable variability in biopsy referral practices based on clinical examination alone and many of the genetic and biomarker tests do not have standardized cutoffs to recommend biopsy. Therefore, to determine whether the tests improve the net health outcome, prospective, comparative data are needed on how the test results are expected to be used versus how they are being used in practice, because of information about the associated effects on outcomes. Many test validation populations have included men with positive digital rectal exam (DRE), PSA level outside of the gray zone (between 3 or 4 ng/mL and 10 ng/mL), or older men for whom the information from the PSA test results are less likely to be informative. African-American geno typical men have a high burden of morbidity and mortality, but have not been well represented in the study populations. It is unclear how to monitor geno typical men with low genetic and biomarker risk scores who continue to have symptoms or high or rising PSA levels. Comparative studies of the many genetic and biomarker tests are lacking, and it is unclear how to use the tests in practice, particularly when the results are contradictory. The evidence is insufficient to determine the effects of the technology on net health outcomes.
In 2013 (validity confirmed 2015), the American Urological Association (AUA) published guidelines for the early detection of prostate cancer:
In the U.S., early detection is driven by prostate specific antigen (PSA) – based screening followed by prostate biopsy for diagnostic confirmation.
While the benefits of PSA-based prostate cancer screening have been evaluated in randomized-controlled trials, the literature supporting the efficacy of DRE, PSA derivatives and isoforms (e.g. free PSA, 2proPSA, prostate health index, hK2, PSA velocity or PSA doubling time) and novel urinary markers and biomarkers (e.g. PCA3) for screening with the goal of reducing prostate cancer mortality provide limited evidence to draw conclusions. While some data suggest use of these secondary screening tools may reduce unnecessary biopsies (i.e reduce harms) while maintain the ability to detect aggressive prostate cancer (i.e. maintain the benefits of PSA screening), more research is needed to confirm this. However, the likelihood of future population-level screening study using these secondary screening approaches is highly unlikely at least in the near future.
In 2013, the Evaluation of Genomic Applications in Practice and Prevention Working Group published the following recommendations for PCA3 testing in prostate cancer, based on the Agency for Healthcare Quality and Research comparative effectiveness review:
In 2015, the National Institute for Health Clinical Excellence (NICE) issued a diagnostic guidance (DG17) regarding diagnosing prostate cancer using Progensa PCA 3 Assay and Prostate Health Index (phi). The NICE recommendation states: The Progensa PCA3 assay and the Prostate Health Index are not recommended for use in people having investigations for suspected prostate cancer, who have had a negative or inconclusive transrectal ultrasound prostate bibopsy.
Indications for Biopsy
Consider biomarkers that improve the specificity of screeningi
Consider multiparametric MRI
iBiomarkers that improve the specificity of detection are not, as yet, recommended as first line screening tests. However, there may be some patients who meet PSA standards for consideration of prostate biopsy, but for whom the patient and/or the physician which to further define the probability of high grade cancer. A percent-free PSA < 10%, PHI > 35 or 4K score (which provides a estimate of the probability of high grade prostate cancer) are potentially informative in patients who have never undergone biopsy or after a negative biopsy; a PCA3 score > 35 is potentially informative after a negative biopsy. The predictive value of the serum biomarkers discussed above has not been correlated with that of MRI. Therefore, it is not known how such tests could be applied in optimal combination.
When the first recommendations for early detection programs for prostate cancer were made, serum tPSA was the only PSA-based test available. PSA derivatives and other assays exist that potentially improve the specificity of testing and thus may diminish the probability of unnecessary biopsies.
When a patient meets the standards for biopsy, sometimes the patient and physician wish to further define the probability of cancer before proceeding to biopsy with is associated risks. Several biomarker tests have been developed with the goals of refining patient selection for biopsies, decreasing unnecessary biopsies, and increasing the specificity of cancer detection, without missing a substantial number of higher grade (Gleason ≥ 7) cancers. These tests may be especially useful in men with PSA levels between 3 and 10 ng/mL. Most often, these tests have been used in patients who have had negative biopsy to determine if repeat biopsy is an appropriate consideration.
The Panel recommends consideration of percent free PSA (%f PSA), 4Kscore, and Prostate Health Index (PHI) and 4Kscore, in patients with PSA levels > 3ng/mL who have not yet had a biopsy. %fPSA, PHI, 4Kscore, PCA3 and ConfirmMDx may also be considered for men who have had at least one prior negative biopsy and are thought to be a higher risk. Results of biomarker assays can be complex and should be interpreted with caution. Referral to specialist should be considered. It should be pointed out that multiparametric MRI is also a consideration in these same patients.
Head-to-head comparisons have been performed in Europe for some of these tests, used independently or in combinations in the initial or repeat biopsy settings, but sample sizes were small and results varied. Therefore, the panel believes that no biomarker test can be recommended over any other at this time. Furthermore, a biomarker assay can be done alone or in addition to multiparametric MRI/refined biopsy techniques in the repeat biopsy setting. The optimal order of biomarker tests and imaging is unknown; and it remains unclear how to interpret results of multiple tests in individual patients – especially when results are contradictory. Results of any of these tests, when performed, should be included in discussions between clinician and patient to assist in decisions regarding whether to proceed with biopsy.
PCA3 is a noncoding, prostate tissue specific RNA that is over-expressed in prostate cancer. Current assays quantify PCA3 over-expression in post-DRE urine specimens. PCA3 appears most useful in determine which patients should undergo repeat biopsy.
The panel believes that this test is not appropriate to use in the initial biopsy setting.
The FDA has approved the PCA3 assay to help decide, along with other factors, whether a repeat biopsy in men age 50 years or older with one or more previous negative prostate biopsies is necessary. This assay is recommended in men with previous negative biopsy in order to avoid repeat biopsy by the Molecular Diagnostic Services Program (MOiDX), and is therefore covered by CMS (Centers for Medicare and Medicaid Services) in this setting.
The Prostate Health Index ([hi) is a combination of the tPSA, fPSA and proPSA tests. The PHI was approved by the FDA in 2012 for use in those with serum PSA values between 4 and 10 ng/mL.
The 4Kscore test is another combination test that measures free and tPSA, human kallikrein 2 (hK2), and intact PSA and also considers age, DRE results, and prior biopsy status. This test reports the percent likelihood of finding high grade (Gleason ≥ 7) cancer on biopsy.
The panel consensus is that the test can be considered for patients prior to biopsy and for those with prior negative biopsy for men thought to be at higher risk for clinically significant prostate cancer. It is important for patients and their urologists to understand, however, that no optimal cut-off threshold has been established for the 4Kscore. If a 4Kscore test is performed, the patient and his urologist should discuss the results to decide whether to proceed with a biopsy.
ConfirmMDx is a tissue based, multiplex epigenetic assay that aims to improve the stratification of men being considered for repeat prostate biopsy. Hypermethylation of the promotor regions of GSTP1, APC, and RASSF1 are assessed in core biopsy tissue samples. The test, performed in on CLIA-certified laboratory, is not FDA approved.
The panel believes that ConfirmMDx can be considered an option for men contemplating repeat biopsy because the assay may identify individuals at higher risk of prostate cancer diagnosis on repeat biopsy. This assay is approved for limited coverage by MoIDX for the reduction of unnecessary repeat prostate biopsies.
The list of assays with the potential to permit improved detection of Gleason Score > 7 prostate cancers as an adjuvant to PSA screening is growing rapidly. Below, several of these assays are discussed. Given the lack of validation of the models/algorithms in additional, independent publications, their unclear behavior in other screened populations, and the lack of clarity regarding the incremental value and cost effectiveness of these assays, however, the panel cannot recommend their routine use at this time. Furthermore, potential sources of error in these approaches include undetectable cancers, as high as 25%, in patients with a single negative prostate biopsy. Other significant and unaddressed issues include the well-known upgrading (32%-49%) that occurs in patients with Gleason 6 cancer at biopsy at the time of pathologic assessment of the surgical specimen. Longer term follow-up of the cohorts to determine whether missed prostate cancers were ultimately detected is needed. In addition, validation of these tests in other cohorts of men is needed before they can be accepted as alternatives to (or perhaps preferable) other tests, described above.
ExoDx Prostate (IntelliScore), also called EPI, evaluates a urine based 3 gene exosome expression assay utilizing PCA3 and ERG (V-ets erythroblastosis virus E26 oncogene homologs) RNA from urine, normalized to SPDEF (SAM pointed domain containing Ets transcription factor). The background for these markers is supported by a number of studies, but the application to exosome detection is unique. This gene panel proposes to discriminate prostate cancer with a Gleason score of ≥ 7 from that of a Gleason score of 6 and benign disease at initial biopsy. The population for which use of this assay was intended to be used includes patients older than 50 years with no prior biopsy and a PSA value between 2 and 10 ng/mL.
The panel considers EPI to be investigational at the present time, but will review additional information as it becomes available.
The Mi-Prostate Score (MiPS) assay measures total serum PSA and post DRE urine expression of PCA3 and the TMPRSS2-ERG fusion gene. Rearrangements of the ERG gene are found in approximately half of prostate cancers. The TMPRTSS2-ERG fusion specifically occurs at high frequency and appears to be an early event in prostate cancer development.
The panel considers MiPS to be investigational at the present time, but will review additional information as it becomes available.
SelectMDx is a gene expression assay performed on post DRE urine that measures DLX1 and HOXC6 expression against KLK3 as internal reference. DLX1 and HOXC6 have been associated with prostate cancer aggressiveness. As with other assays, SelectMDx is designed to improve the identification of men with clinically significant prostate cancer prior to biopsy, thereby reducing the number of unnecessary biopsies.
The panel considers SelectMDx to be investigational at the present time, but will review additional information as it becomes available.
The current NCCN guideline does not include or indicate the use of APIFINY biomarker testing for prostate cancer risk assessment or management.
The U.S. Preventative Services Task Force (USPSTF) published an updated recommendation in 2018 for prostate cancer screening, and the recommendation does not address genetic or protein biomarker testing.
See Related Medical Policy
Genetic and protein biomarkers for the diagnosis of prostate cancer are considered investigational. This includes, but is not limited to the following:
To date, the majority of available studies fail to provide sufficient evidence that genetic and protein biomarker testing for the diagnosis and cancer risk assessment of prostate cancer lead to improved health outcomes or change in management treatment decisions (i.e. clinical utility). Well - designed randomized controlled trials (RCTs) are needed to determine the clinical utility of genetic and protein biomarker testing for the diagnosis and cancer risk assessment of prostate cancer compared to traditional clinical factors/testing to guide medical management and improve clinical outcomes. The evidence is insufficient to determine the effects of this testing on net health outcomes.
Single nucleotide variant (SNVs) testing for cancer risk assessment of prostate cancer is considered investigational.
Numerous studies have demonstrated the association of many gene panels and single nucleotide variant testing (SNVs) with prostate cancer. These studies, in early stages of development, have generally shown a modest degree of association with future risk for prostate cancer. The clinical utility of these tests is uncertain; there is no evidence that information obtained from gene panels and SNV testing can be used to change management in ways that improve outcomes. Therefore, gene panels and SNV testing for cancer risk assessment of prostate cancer is considered investigational.
To report provider services, use appropriate CPT* codes, Alpha Numeric (HCPCS level 2) codes, Revenue codes and / or diagnosis codes.
Wellmark medical policies address the complex issue of technology assessment of new and emerging treatments, devices, drugs, etc. They are developed to assist in administering plan benefits and constitute neither offers of coverage nor medical advice. Wellmark medical policies contain only a partial, general description of plan or program benefits and do not constitute a contract. Wellmark does not provide health care services and, therefore, cannot guarantee any results or outcomes. Participating providers are independent contractors in private practice and are neither employees nor agents of Wellmark or its affiliates. Treating providers are solely responsible for medical advice and treatment of members. Our medical policies may be updated and therefore are subject to change without notice.
*CPT® is a registered trademark of the American Medical Association.